Scientific Sense Making in Context

• In Event: Activating Young Science Learners: Igniting Persistent Engagement in Science Learning and Inquiry

Abstract

The objective of this poster session is to present findings from research and development related to measuring scientific sensemaking in context. We take the view that scientific sensemaking is situated fundamentally in seeking and applying cohesive evidence to advance justificatory and comparative arguments (Berland and McNeill, 2010). The theoretical framework for this study builds on the work of the Science Learning Activation Lab (Dorph, et. al., 2011). This framework posits that activation in science consists of four discrete and overlapping areas that include: fascination by natural and physical phenomenon, perseverance in scientific endeavors, valuing science, and engagement in scientific sensemaking.

The study and measure presented in this session focuses on two of these aspects; valuing science and engagement in scientific sensemaking. As such, our measure includes evidence, explanations, generating investigable questions, mechanistic reasoning, coherence seeking, and nature of science. Recognizing that effort and motivation are heavily impacted by interest (Jenkins & Pell, 2006; Planty et al., 2009; Bybee, McCrae, & Laurie, 2009), the instrument is situated across a number of content topics selected because they were identified as “favorites” of 11-year-olds. Students choose from these topics and progress through the measure within their chosen topic.

Data collected first included a pilot study of 40 students who took the assessment one time and 10 students who participated in cognitive lab interviews to support measure validation efforts. Next, a larger scale study was conducted with 150 demographically diverse 11 and 12 year olds. Students completed the measure described above twice in both formal and informal environments. Item Response Modeling was used to investigate items and student abilities from the two administrations. A Partial Credit Rasch model (Masters, 1982) was fit to the data using ConQuest Generalized Item Response Modeling software (Wu, Adams, Wilson & Haldane, 2007). Student response data was used to estimate student abilities and item difficulties on the posttest instrument and item difficulties were anchored in the pretest.

Preliminary results showed that the items fit the model and were able to distinguish a range of student abilities in scientific sensemaking. Changes in student ability estimates showed improvement in students’ generation of causal and mechanistic explanations and the inclusion of evidence in their explanations. Initial findings suggest that for the students with the highest abilities, the role of content is less important than for those students with middle and lower level abilities. Further, these results suggest that the sophistication of scientific sensemaking in middle school children can be positively affected by a range of learning environments.

These findings are of significance as they show alignment with the theoretical developmental trajectories of scientific sensemaking defined in this work. Use of this measure in the field will support advances in scholarship and practice. From a research perspective, the measure supports further studies related to the development of scientific sensemaking. From a practice perspective, this measure offers a valid option for measuring an outcome that is of great interest to practicioners and evaluators of science learning across multiple settings.

Authors

• Kristin Nagy Catz, University of California - Berkeley

• Amanda Crowell, University of Pittsburgh

• Kristen Orourke Burmester, University of California - Berkeley

• Christian D. Schunn, University of Pittsburgh

• Rena Dorph, University of California - Berkeley