[Originally published as What Are Science Models?]
Defining a Science Model
A science model is: a testable idea used to describe a phenomenon. Models are based on a set of observations and are made by finding patterns in what is observed. Generally, because a model works with patterns, it is capable of being applied to many similar situations. Science models need to be testable, falsifiable, or able to be disproved in some way.
A useful model is one that can explain many aspects of a phenomenon in a simple way. The more detail and complexity that is placed into a model the more powerful it becomes at accurately predicting events. However, adding too much complexity to a model can make it difficult to work with. Therefore, most models are a compromise between their power to predict and their simplicity of use.
Any model that is too difficult to use or too simple to explain the phenomena is not a useful model.
Models can come in a symbolic form rather than a natural form. A mathematical formula is a powerful symbolic model. Various details of phenomena can be represented symbolically and then tested by simply changing the value of those symbols.
With the extensive use of such models in recent centuries, many phenomena have already been expressed by a mathematical symbol or formula adding to the ease of use for future models. Mathematical models often use the field of statistics and discuss predictions in terms of probabilities.