Much problem solving research to this point has focused on differences between expert and novice problem solvers. Only a few studies have made any effort to describe how the transition from novice to expert occurs. Also, most problem solving studies have been set in laboratories, few in actual classrooms. In an effort to learn more about these unexplored areas, a series of studies was designed.

One portion tracked the development of physics problem solving skills with documented differences between experts and novices. These included the drawing and use of physical representations, writing general or specific initial equations, writing algebraic or numerical initial equations, insertion of numbers, word usage, and fractionation. The evolution of these skills was followed by examining written exam data. A small probe was made for indications of potential transfer by examining a final exam problem from an engineering statics course taken by a small portion of the initial research sample.

To explore students in the transitional state, a series of interview tasks was designed. Two tasks, also administered to faculty for comparison, involved evaluation of previously written solutions. The first provided subjects with a student solution and asked whether the solution was a good one or not. The second asked subjects to classify solutions written by students or instructors. A third task asked students to solve two complex electricity and magnetism problems in a think-aloud interview. This task was designed to check validity of the methods developed to analyze the written exam solutions.

This research was carried out in a two-quarter introductory physics sequence for honors engineering freshmen. The purpose was not to evaluate the instruction of the course, but to describe student development in this environment. The interviews and probe into the transfer issue took place the following quarter. Qualitative and quantitative methods were used to analyze the data.