Journal of Science Education and Technology

, Volume 25, Issue 5, pp 806–823

Order Matters: Sequencing Scale-Realistic Versus Simplified Models to Improve Science Learning


DOI: 10.1007/s10956-016-9642-4

Cite this article as:
Chen, C., Schneps, M.H. & Sonnert, G. J Sci Educ Technol (2016) 25: 806. doi:10.1007/s10956-016-9642-4


Teachers choosing between different models to facilitate students’ understanding of an abstract system must decide whether to adopt a model that is simplified and striking or one that is realistic and complex. Only recently have instructional technologies enabled teachers and learners to change presentations swiftly and to provide for learning based on multiple models, thus giving rise to questions about the order of presentation. Using disjoint individual growth modeling to examine the learning of astronomical concepts using a simulation of the solar system on tablets for 152 high school students (age 15), the authors detect both a model effect and an order effect in the use of the Orrery, a simplified model that exaggerates the scale relationships, and the True-to-scale, a proportional model that more accurately represents the realistic scale relationships. Specifically, earlier exposure to the simplified model resulted in diminution of the conceptual gain from the subsequent realistic model, but the realistic model did not impede learning from the following simplified model.


Order effect Model-based learning Misconception STEM education 

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.Harvard Graduate School of EducationCambridgeUSA
  2. 2.University of MassachusettsBostonUSA
  3. 3.Massachusetts Institute of TechnologyCambridgeUSA
  4. 4.Harvard-Smithonian Center for AstrophysicsCambridgeUSA

Personalised recommendations