Sedimenting Science Practices: Longitudinal Analysis of Modeling Practices and Artifacts

• In Event: 63.025 - Supporting Modeling Epistemologies in the Science Classroom

Abstract

Objectives. International reform efforts and the NGSS have focused increasingly on science practices, such as modeling (Duschl, Schweingruber & Shouse, 2007; Erduran & Jiménez-Aleixandre, 2007; NRC, 2012; Östman & Wickman, 2014). These practices are difficult to support and evaluate, because students’ enactments are context-dependent (e.g. Author, 2017; Ford, 2015; Manz, 2014; Rouse, 2007). Thus, it is important to understand how contextual features shape classroom norms and individual practices. We use a lens of sedimenting to consider the ways in which norms, practices, and artifacts are substrates for students’ later actions (Goodwin, 2017). We aim to illustrate the dialectic and accumulative relationship between these substrates with data from a longitudinal design study.

Theoretical Framing. As students engage in modeling, they generate practices, epistemic goals, and cultural and material resources across several timescales and meaning systems (Lemke 1995; 1999; Pahl, 2003). These emergent resources become raw materials for new and complex patterns (Lemke, 2000). Therefore, it is important to consider the historical context of students’ models and practices. Goodwin (2017, p. 32) describes these accrued resources as substrates, capitalizing on the metaphor that a “substrate is both the sedimented outcome of earlier action, and the source of subsequent action.” From this perspective, it is important to consider how substrates become sedimented resources, what gets sedimented, and how substrates are appropriated or disrupted (Keifert & Marin, 2018), particularly with socially constructed practices like modeling.

Methods/Data. We partnered with an 8th grade teacher and his 91 students to conduct an iterative design-based classroom experiment (Cobb, Confrey, diSessa, Lehrer, & Schauble, 2003). Data include students’ artifacts (models, reflections, interviews, and assessments) and video and audio of each class session. Initially, we organized our coding categories around Schwarz and colleagues’ (2012) respected learning progression for modeling (Authors, 2017). New themes emerged in our analysis related to the role of students’ prior models and experiences. We expanded and refined our codes through constant-comparative analysis (Strauss & Corbin, 1990). We traced the historical narrative of students’ modeling practices, analyzing the role of prior artifacts and experiences as substrates for later action.

Results and Scholarly Significance. We illustrate several of the ways that substrates sediment into classroom practices with three episodes (see Table 1 for examples). First, we show how the teacher’s instruction was an immediate substrate for anticipated and emergent student practices. Second, we describe how historical tangible and intangible substrates influence students’ later models, again in anticipated ways and emergent ways. Third, we demonstrate that students’ previous modeling practices are substrates that both support and constrain subsequent practices.
Our findings are significant because they show that a historical dimension of analysis is critical for understanding students’ modeling practices. Because students’ previous practices and artifacts are substrates, the affordances and constraints of each learning activity carry over into students’ subsequent practices. Therefore, researchers and educators should consider implications for students’ future modeling beliefs and practices when designing learning activities.

Authors

• Ashlyn Pierson, Vanderbilt University

• Doug Clark, University of Calgary