This paper discusses some means by which one might answer the question of "What makes problem solving hard?" In order to facilitate the investigations required to answer this question, a model of problem solving is needed. From some simulated solutions to a physics problem, several problem solving processes used by students are identified. A concise model of problem solving based on one of these solutions is then described. The model identifies steps that must logically be worked to complete the problem. These steps include: (1) reading; (2) sketching: translating words into symbols; (3) recognizing the law governing situations; (4) identifying definitions for quantities found in the law; (5) solving equations; and (6) making numerical substitutions (simplification skills). Limits to the validity of the model are discussed in light of the other solutions. Finally, an instrument that can be constructed by means of the model is described and its reliability tested. Subsequent papers will report on the results of applying the model to the learning of physics in a typical introductory course. (Author/JN)