An important goal of any physics course is student understanding of the key concepts. Another goal is to provide the students with the tools needed to do the physics. Pragmatic concerns make it difficult to achieve both of these goals. Many physics courses overemphasize theory, formal manipulations, and problem solving, and do not sufficiently explore related aspects of experimental and computational physics. The de-emphasis of experimental physics is partly due to the large costs both in money and time. The relative lack of computational physics is due to inertia, the amount of beautiful theory that we wish to show our students, and the time commitment involved in developing appropriate educational materials and in student time. Also, the time spent writing, modifying, and using computer programs can be excessive. Because much of statistical physics can naturally be expressed in terms of computer algorithms and because there are very few exact results accessible to undergraduates, a course in statistical physics is a natural one for incorporating a significant computational component. We will argue that not only is a computational approach valuable, if not necessary, for an understanding of the concepts in statistical physics, but it also provides an introduction to the tools we use in research. We will present an approach that is pedagogically motivated as well as practical.
展开▼
