Search

The recent advancement of large language models presents numerous opportunities for teaching and learning. Despite widespread public debate regarding the use of large language models, empirical research on their opportunities and risks in education remains limited. In this work, we demonstrate the qualities and shortcomings of using ChatGPT 3.5 for physics task development by prospective teachers. In a randomized controlled trial, 26 prospective physics teacher students were divided into two groups: the first group used ChatGPT 3.5 to develop text-based physics tasks for four different concepts in the field of kinematics for 10th grade high school students, while the second group used a classical textbook to create tasks for the same concepts and target group. The results indicate no difference in task correctness, but students using the textbook achieved a higher clarity and more frequently embedded their questions in a meaningful context. Both groups adapted the level of task difficulty easily to the target group but struggled strongly with sufficient task specificity, i.e., relevant information to solve the tasks were missing. Students using ChatGPT for problem posing rated high system usability but experienced difficulties with output quality. These results provide insights into the opportunities and pitfalls of using large language models in education., Comment: 10 pages, 3 figures, 3 tables

Optics is a core field in the curricula of secondary physics education. In this study, we present the development and validation of a test instrument in the field of optics, the Converging Lenses Concept Inventory (CLCI). It can be used as a formative or a summative assessment of middle school students' conceptual understanding of image formation by converging lenses. The CLCI assesses: (1) the overall understanding of fundamental concepts related to converging lenses, (2) the understanding of specific concepts, and (3) students' propensity for difficulties within this topic. The initial CLCI consists of 16 multiple-choice items; however, one item was removed based on various quality checks. We validated the CLCI thoroughly with distractor analyses, classical test theory, item response theory, structural analyses, and analyses of students' total scores at different measurement points as quantitative approaches, as well as student interviews and an expert survey as qualitative approaches. The quantitative analyses are mostly based on a dataset of N = 318 middle school students who took the CLCI as a posttest. The student interviews were conducted with seven middle school students after they were taught the concepts of converging lenses. The expert survey included five experts who evaluated both individual items and the test as a whole. The analyses showed good to excellent results for the test instrument, corroborating the 15-item CLCI's validity and its compliance with the three foci outlined above.

The concepts of the Coriolis and the centrifugal force are essential in various scientific fields and they are standard components of introductory physics lectures. In this paper we explore how students understand and apply concepts of rotating frames of reference in the context of an exemplary lecture demonstration experiment. We found in a $Predict-Observe-Explain$-setting, that after predicting the outcome prior to the demonstration, only one out of five physics students correctly reported the observation of the trajectory of a sphere rolling over a rotating disc. Despite this low score, a detailed analysis of distractors revealed significant conceptual learning during the observation of the experiment. In this context, we identified three main misconceptions and learning difficulties. First, the centrifugal force is only required to describe the trajectory if the object is coupled to the rotating system. Second, inertial forces cause a reaction of an object on which they act. And third, students systematically mix-up the trajectories in the stationary and the rotating frame of reference. Furthermore, we captured students' eye movements during the $Predict$ task and found that physics students with low confidence ratings focused longer on relevant task areas than confident students despite having a comparable score. Consequently, this metric is a helpful tool for the identification of misconceptions using eye tracking. Overall, the results help to understand the complexity of concept learning from demonstration experiments and provide important implications for instructional design of introductions to rotating frames of reference., Comment: 12 pages