Statement on Computational Physics

The American Association of Physics Teachers urges that every physics and astronomy department provide its majors and potential majors with appropriate instruction in computational physics.
Contemporary research in physics and related sciences almost always involves the use of computers.  They are used for data collection and analysis, numerical analysis, simulations, and symbolic manipulation. Computational physics has become a third way of doing physics and complements traditional modes of theoretical and experimental physics.  In addition, almost all undergraduate students who take physics courses will use computational tools in their future careers even if they do not become practicing physicists.
One of the traits that distinguishes physics from other sciences is our ability to develop new tools as needed to do our work.  These new tools include new experimental techniques, mathematical methods, theoretical formalisms, and now new computer algorithms.  Thus, we should include in the physics undergraduate curriculum some level of instruction in computer algorithms appropriate for solving problems in physics.
Insight into understanding physics can be gained in many ways.  Experiments emphasize that our models are connected to the real world, and frequently surprise us with new phenomena we didn't expect.  Theory provides the tools for organizing our knowledge, making predictions, and describing nature in a concise and compelling manner.   The computer provides a new tool that enhances both theory and experiment. Computer simulations allow us to develop models that are not solvable analytically, to test theories where traditional experiments are too difficult or expensive, to ask “what-if” questions, and to visualize the time development of dynamical systems. As a result simulations provide different insights, which may not be possible to obtain through the use of traditional theoretical and experimental methods.


Approved by the AAPT Executive Board at the Spring 2011 meeting.