Longitudinal Analysis of Upper Elementary and Middle School Students' Modeling Practices

• In Event: 32.029 - Supporting Science as a Modeling Practice in the Classroom Through the Lens of Next Generation Science Standards

Abstract

Objectives. Our research presents longitudinal analysis of five students’ modeling work across two years (5th and 6th grades) within classrooms that supported scientific practices. Our goal was to determine how students’ modeling practices developed over time and what supported that development.
Theoretical Framework. Our theoretical framework is based on work in modeling education including our learning progression of scientific modeling (Authors, 2009) and our subsequent work outlining epistemologies in practice (EiP framework) (Authors, 2015). In particular, we analyzed four epistemic considerations as students developed, evaluated and revised model-based explanations (MBE) of how/why phenomena occur. Those included what kind of answer the MBE should provide (nature), how the MBE relates to other scientific phenomena and ideas (generality), how students justify ideas in the MBE, and who will use the MBE and how (audience).
Methods and Data. We analyzed data from five focal students over two years. These case students were chosen from a larger sample of participants (N=200) and focal students (N=25) for their complete information, range of gender and ethnicity, and detailed responses to interview questions. We analyzed nine interviews with each focus student about their MBEs across two years as well as written assessments and diagrammatic models from their classroom work. We coded students’ interviews and work based on EIP considerations and emphasis over time.
Results. Longitudinal analysis indicates that students were unique in their modeling development over time and incorporated particular affinities for various epistemic considerations that sustained them in their work. For example, Lily leveraged the idea that a model-based explanation should answer ‘how and why” to advance her understanding of nature and generality across years; Jessie brought her careful attention to detail and evidence to develop causal mechanisms in her models (nature). Shanti and Rudo leveraged the importance of justifying ideas and addressing ideas from an audience; Miriam enjoyed hands-on activities and did not verbally identify a particular epistemic consideration in her modeling work.
Analysis of interviews also indicated aspects of the learning environment that supported modeling. One fifth-grade teacher who emphasized that models should address how and why something happens (nature), and be applied to other contexts (generality) powerfully influenced several students (e.g., Lily, Jessie). The curriculum and teaching environments that supported peer evaluation and consensus were also powerful for showing students that peers have ideas that can be informative (audience), and that evidence (proof, empirical, etc.) is important for supporting claims (justification). Further, having an audience with competing or coexisting ideas, and forming a consensus model can be highly motivating for students’ model development (e.g., Shanti, Rudo).
Scholarly significance. In-depth longitudinal work with students engaged in scientific modeling is rare and important for gaining insights in to how modeling can develop in school settings and how to best support it through research and teaching. Our analysis, focusing on EIPs about students’ own work, as well as our findings about the benefits of leveraging EiPs for teachers and students can help advance the field and support of scientific modeling in classrooms.

Authors

• Christina V. Schwarz, Michigan State University

• Li Ke, Michigan State University