A model is an abstraction of a phenomenon that represents selected aspects of their relationships in order to explain, describe, predict, or investigate it. Scientists develop models and simulations to decode complex ideas and phenomena across a wide range of disciplines, such as economics, mathematics, physics, meteorology, biology, and the social sciences (for example, models that describe changes in Earth’s temperature in relation to greenhouse gas emissions and enable predictions regarding climate change).
The process of modeling, within which a model is constructed and evaluated, is a process of high educational value, not only for scientists, but also for students. In modeling processes, students are responsible for defining the phenomenon and the questions they seek to investigate (for example, when modeling the water cycle, learners may focus on the evaporation process); for identifying and defining the factors involved in the phenomenon (for example, in modeling the water cycle, learners need to identify factors such as the amount of radiation and its effect on evaporation); for mapping the relationships and mutual influences among these factors; and for interpreting the model they created, evaluating it, and considering possible ways to improve it. At the end of the process, learners are expected to explain the phenomenon and to generalize what they learned about it through the modeling process. This can be done as a thought experiment, using paper and pencil, or through dedicated computer-based tools.
Modeling activities can support deep understanding of the phenomena under study and, beyond that, the development of thinking skills such as abstraction, generalization, and distinguishing between essential and non-essential elements. Modeling activities can be combined with more direct approaches (such as experimentation) for investigating complex phenomena. Comparing the phenomenon as observed in an experiment and as represented in a model enables learners to better understand the phenomena and to refine both the model and the experiment.
An example of modeling processes in citizen science is the “Radon Gas” Project, in which students engage in a range of modeling processes (primarily statistical modeling) using the technological tool TinkerPlots in order to learn about the phenomenon of radon gas. These modeling processes enable students to deepen their understanding of complex statistical ideas such as data, samples and sampling, inference, and more.
Deepening and Expansion ▼
What is a model?
According to Lehrer and Schauble (2010), a model is a type of explanation that characterizes, and even defines, science. A model can be defined as an analogy that represents selected objects from the phenomenon under study and the relationships among them (for example, billiard-ball models simulating the behavior of gases, or models of the solar system), with the aim of explaining a particular phenomenon in a particular context. The model provides an abstraction of the phenomenon. A model may be abstract (conceptual) or tangible (a physical model, a graphical representation, or a representation in a technological tool). The purposes of model construction may include describing a phenomenon, explaining it, predicting its behavior, and more. It is important to note that if an analogy or representation was not developed for a specific purpose, it would not be considered a model.
What is modeling? And what is model-based thinking?
According to Lehrer and Schauble (2010), the process of thinking and constructing models is referred to as modeling. This process includes examining, evaluating, and improving models in order to create an analogy that best represents the phenomenon under study, in accordance with the purpose for which the model was constructed. The modeling process involves examining theoretical ideas and findings and integrating them into the model being constructed or refined.
The thinking involved in modeling (model-based thinking) is a complex process
Lehrer and Schauble (2010) explain this complexity as stemming from the fact that modeling involves moving from real-world objects and the relationships among them toward constructing a representation that stands in for them. This alternative representation highlights aspects of the phenomenon that are perceived as important. The modeling process enables indirect learning about the world through a model in which the phenomenon is presented in a simplified manner.
Lehrer and Romberg (1996) describe model-based thinking as a bridge that enables movement from intuitive understandings toward scientific understandings of the world. When students engage in modeling, they generate data about the phenomenon under study through the models they construct, while simultaneously examining the data to better understand the phenomenon. This process involves creating, testing, and improving models and supports the development of understanding of the phenomenon under study. The thinking involved in modeling processes is therefore considered a practice characteristic of scientific research.
The contribution of model-based thinking to scientific thinking
According to Lesh and Lehrer (2003), scientific inquiry is inherently involved in the design, construction, and refinement of models of the world. Researchers’ ideas may emerge from investigating models, and scientific theories may change through the process of building and refining models and through comparison of multiple models.
Engagement in modeling is significant in scientific inquiry because it fundamentally involves posing questions, investigating them, drawing conclusions, engaging in critical thinking, and searching for new ideas and questions. Students who engage in modeling learn to generate, evaluate, and investigate meaningful questions of their own. They ask what counts as a good question and what constitutes good evidence, and they learn that the goal of scientific inquiry is not to find the “correct” answer (for example, one already known by the teacher). Rather, discovery inevitably leads to further questions, and sustained investigation of a phenomenon may lead to deep understanding. Emphasis on modeling can therefore lead to shifts in students’ conceptions of what knowledge is, and in particular, what scientific knowledge is.
Cultivating model-based thinking
In a chapter on model-based thinking in scientific contexts, Lesh and Lehrer (2003) ask how modeling practices can be cultivated in ways that build on children’s developing capabilities and enable them to participate in scientific activity that includes inventing and improving models of phenomena in the world. Their point of departure is that a model is an analogy; therefore, to cultivate thinking about models in ways that build on children’s developing capabilities, they examine how thinking about analogies develops. The literature describes a progression of complexity in analogies, expressed in the mapping relationship between the model and its source. Early structures may range from verbal or physical resemblance to the phenomenon (a replica) to fully relational structures. The desire to construct a model with physical similarity is reinforced by the social context, as it is easier to persuade others that a model represents a phenomenon if it physically resembles the phenomenon represented. Accordingly, the authors propose a classification of models characterized by the intentions and goals of the modelers rather than solely by the qualities of the models.
This mapping, which describes four types of models, also serves as a proposal for how models can be introduced to students in a gradual manner:
- Physical microcosms:
Models that derive their power from physical similarity and direct correspondence. These models are miniature representations (either smaller than the phenomenon, such as Earth, or larger, such as a representation of a cell) of the systems they represent. They may include relationships among the structures described within a given context. Physical models provide entry into modeling through literal resemblance while also enabling questions about relational structure, such as: What should be included in the model, and why? How should it be modified to incorporate new observations and data? - Models as representational systems:
Systems of scientific representations (records), such as maps of Earth. These models can be easily transported, modified, and recreated while preserving relationships among aspects of the represented phenomenon. They support thinking about representation itself, for example, how one thing represents another. Such systems both express and shape model-based thinking by enabling the use of records, symbols, and understandings of their contributions, which is fundamental to the modeling process. - Syntactic models:
These models summarize the primary functioning of a system and typically bear little physical resemblance to the system they represent. Syntactic models derive their power from relational mapping—that is, mapping that preserves relationships (for example, prediction using a dice roll). - Emergent models:
In these models, links between the model and the real world arise from relationships among entities that generate emergent behavior not previously visible in earlier descriptions (for example, the outbreak of a disease).
References ▼
Kali, Y. (2006). Collaborative knowledge-building using the Design Principles Database. International Journal of Computer-Supported Collaborative Learning, 1(2), 187–201.
Lehrer, R., & Romberg, T. A. (1996). Exploring children's data modeling. Cognition and Instruction, 14(1), 69–108.
Lehrer, R., & Schauble, L. (2010). What kind of explanation is a model? In M. K. Stein (Ed.), Instructional Explanations in the Disciplines (pp. 9–22). New York: Springer.