A scientific model is an abstraction or a simplification of a system that makes its central features explicit and visible (Harrison & Treagust, 2000). Scientists use models as thinking tools; this practice includes helping to develop research questions and explanations and helping to test predictions. Scientists also use models as tools to help them communicate their ideas with others; this practice also often involves the articulation of underlying causal mechanisms for phenomena.

Scientific modeling can be decomposed into several component practices: constructing, using, evaluating, and revising models (Schwarz et al., 2009). These component practices are further elaborated below:

  • Constructing models: One can use a hypothesis informed by principles or theories to construct models, and apply appropriate principles, theories, or evidence.

  • Using models: One can use multiple models to explain and predict aspects of a group of related phenomena.

  • Evaluating models: One can consider the predictive or explanatory power of a model and determine whether a model makes accurate and reliable predictions, whether it accounts for all available evidence, and/or whether there is consistency with all available evidence.

  • Revising models: One can revise a model when the predictive power or explanatory power is called into question, or when new evidence is available that challenges predictive or explanatory power.

We expect that students may have some experience constructing models(particularly physical models) in classrooms; however, few students may have experience using, testing, or revising models. Therefore, when assessing students’ modeling skill, it is necessary to provide some examples of how scientists use, test, and revise their models in order to demonstrate and scaffold this practice for students. In addition, being proficient in modeling includes conscious separation of a model and its referent, as well as the explicit consideration of measurement error and consideration of alternative models (Lehrer & Schauble, 2000). Therefore, when assessing students’ modeling skills, it is also important to prompt them to consider alternative models and limitations of models. For these reasons, in our CBAL tasks we scaffold students’ ability to reflect on scientific models (i.e. their epistemic knowledge) and the practice of modeling.

Back to Science Practices

Home | About CBAL | Acknowledgments | Contact Us

© Copyright 2014 Educational Testing Service. All rights reserved.

ETS and the ETS logo are registered trademarks of Educational Testing Service (ETS). CBAL is a trademark of ETS.
All other trademarks are the property of their respective owners.