Seeing the Forest for the Trees: An Exploration of Student Problem Solving and Reasoning with 1H NMR Spectral Features
Abstract
Nuclear magnetic resonance (NMR) spectroscopy is vital to synthesis and provides rich problem-solving opportunities for organic chemistry students. However, little is known about 1H NMR spectroscopy instruction or how students use spectral features in solving. The goal of this dissertation research was to examine how students learn about and solve 1H NMR spectroscopy problems. Organic chemistry textbooks were analyzed for the ways in which spectral features were introduced and incorporated into worked examples and practice problems. Spectral features like the number of signals and chemical shift were covered by problems more frequently, while integration was covered least. Think-aloud interviews were completed to identify the operators students utilized in their problem-solving processes, and extra credit problem sets were designed and administered to students at three different universities to examine whether students could correctly perform each individual type of operator. While students could perform operators, it was unclear if students knew how and when to use the operators. To fill this knowledge gap, multiple choice assessment questions were developed and administered to students at three different large universities. Coding schemes were developed to identify and describe students’ use of task features and inferences, and regression analyses were completed to discern which areas of reasoning led to success in solving. A majority of students did not identify using any critical spectral features in written explanations. Regression analyses revealed that the inferences students made, and not the task features they paid attention to, were most significantly associated with success in structural predictions; a majority of students made solely correct inferences in their reasoning explanations. When a mixture of correct and incorrect inferences were made, a majority of those students were unable to answer the questions correctly. These findings suggest that students may know enough to solve simple 1H NMR spectroscopy problems, but may lack knowledge about specific spectral features which could impact overall solving success. Students may require considerable support in deciphering the critical features in 1H NMR spectroscopy problems and developing robust, correct inferences across all spectral features.