Using self-made drawings to support modelling in science education


  • Frank Leenaars studied Computer Science (B.Sc.) and Psychology (B.Sc., M.Sc.) and is currently a PhD student at the Department of Instructional Technology at the University of Twente. He works on the development and evaluation of innovative electronic learning environments for primary school science education using an approach based on drawing and modelling. Wouter van Joolingen is professor of dynamical modelling in educational settings. His work concerns the use of simulations and modelling in science education. His main interest is making computer modelling accessible to young children with the help of intuitive means of model specification, such as drawings and qualitative models. Lars Bollen received his PhD in 2009 at the University Duisburg-Essen on “Activity Structuring and Activity Monitoring in Heterogeneous Learning Scenarios with Mobile Devices.” Currently, he works at the University of Twente on the EU project SCY and on the project “Thinking with Inaccurate Drawings.” His research interests include model-based learning & sketching, pen-based devices in learning scenarios and (inter)action analysis.

Mr. Frank Leenaars, University of Twente (GW/IST), PO Box 217, 7500 AE Enschede, The Netherlands. Email:



The value of modelling in science education is evident, both from scientific practice and from theories of learning. However, students find modelling difficult and need support. This study investigates how self-made drawings could be used to support the modelling process. An experiment with undergraduate students (n = 37) at a predominantly technical university led to three conclusions. 1. Most learners created realistic rather than schematic drawings of real world systems. Furthermore, learners who represented situations realistically identified a greater number of important aspects of these situations than learners who represented them purely schematically. 2. Access to simulations during the construction of these drawings led to increased insight into the effects of variables that can be manipulated. However, participants with access to simulations thought of fewer important variables that were not explicitly available in the simulation than participants without this access. 3. Participants almost never drew multiple objects with a single stroke and generally drew objects sequentially. These patterns in the digital drawing process can simplify automatic sketch segmentation, which can be used to support learners in creating models from drawings.

Practitioner Notes

What is already known about this topic

  • • Modelling plays a key role in scientific discovery and reasoning.
  • • Computer modelling in science education allows for a focus on the processes as well as the products of science.
  • • Students encounter difficulties during the modelling process and need to be supported.

What this paper adds

  • • Self-made drawings could be a good way to support the modelling process, because creating an informal drawing allows students to use their prior domain knowledge more easily than creating a formal model.
  • • Students who created a realistic drawing showed more prior knowledge activation than students who created an abstract drawing.
  • • Compared to students who only had access to a text, students who had access to a simulation during the creation of a drawing showed more insight into some variables of the system they represented, but found it more difficult to think of variables not adjustable in the simulation.

Implications for practice and/or policy

  • • Before students are asked to create a formal model of some domain, their prior domain knowledge may be activated by the creation of an informal drawing.
  • • When asking students to study a domain using a simulation, attention must be paid to the prevention of a focusing effect. This focusing effect may be attenuated by asking students to express their prior domain knowledge in a drawing before using the simulation.