Investigating scientific modeling practices in U.S. and German elementary science classrooms: A comparative, cross‐national video study

Abstract

AbstractScience education reform efforts have emphasized the incorporation of more authentic and meaningful scientific practices into formal learning environments. With this focus on practices, such as scientific modeling, a powerful tool for sense‐making and reasoning, comes a need to observe and understand if and how it is actually occurring, particularly in elementary science learning environments. To address this need, and moving forward an international, collaborative research effort centered around scientific modeling, we undertook a cross‐national observational study in which we used an observation protocol to characterize scientific modeling practices in elementary classrooms in both the United States and Germany. Drawing on existing video‐based datasets from multiple projects, this study's findings provide empirical evidence for model‐based teaching and learning and explore the relative frequencies of relevant modeling subpractices (i.e., construction, use, evaluation, and revision of scientific models) in regular, day‐to‐day elementary science classrooms. This study also illustrates the differences and similarities between German and American science instruction. In general, students in both countries experience model construction more often than other modeling practices. German teachers more consistently foreground anchoring phenomena and students’ questions/preconceptions as part of the modeling process, thus leading to slightly higher overall modeling scores—despite a complete lack of model evaluation and revision in any of the observed German classrooms. Ultimately, however, young learners in both nations seem to experience alarmingly little opportunities to engage in any of the modeling subpractices. Particularly in terms of modeling as a means to make sense of phenomena, constrained by teachers most commonly using finished representations (i.e., models) of phenomena and processes to teach science/deliver content. Our results indicate the need for a greater focus on all scientific modeling subprocesses, particularly the integration of scientific modeling with already established, more general (e.g., inquiry‐based) strategies for effective science teaching and an emphasis on actualizing modeling as an authentic sense‐making endeavor.