(Link) Student learning in upper-division physics courses is a growing area of research in the field of Physics Education. Developing effective new curricular materials and pedagogical techniques to improve student learning in upper-division courses requires knowledge of both what material students struggle with and what curricular approaches help to overcome these struggles. To facilitate the course transformation process for one specific content area — upper-division electrostatics — this thesis presents two new methodological tools: (1) an analytical framework designed to investigate students' struggles with the advanced physics content and mathematically sophisticated tools/techniques required at the junior and senior level, and (2) a new multiple-response conceptual assessment designed to measure student learning and assess the effectiveness of different curricular approaches.
We first describe the development and theoretical grounding of a new analytical framework designed to characterize how students use mathematical tools and techniques during physics problem solving. We apply this framework to investigate student difficulties with three specific mathematical tools used in upper-division electrostatics: multivariable integration in the context of Coulomb's law, the Dirac delta function in the context of expressing volume charge densities, and separation of variables as a technique to solve Laplace's equation. We find a number of common themes in students' difficulties around these mathematical tools including: recognizing when a particular mathematical tool is appropriate for a given physics problem, mapping between the specific physical context and the formal mathematical structures, and reflecting spontaneously on the solution to a physics problem to gain physical insight or ensure consistency with expected results.
We then describe the development of a novel, multiple-response version of an existing conceptual assessment in upper-division electrostatics courses. The goal of this new version is to provide an easily-graded electrostatics assessment that can potentially be implemented to investigate student learning on a large scale. We show that student performance on the new multiple-response version exhibits a significant degree of consistency with performance on the free-response version, and that it continues to provide significant insight into student reasoning and student difficulties. Moreover, we demonstrate that the new assessment is both valid and reliable using data from upper-division physics students at multiple institutions. Overall, the work described in this thesis represents a significant contribution to the methodological tools available to researchers and instructors interested in improving student learning at the upper-division level.