Understanding students' misconceptions and how they change is an essential part of supporting students in their science learning. This paper presents results from distractor-driven multiple-choice assessments that target students' misconceptions about energy. Over 20,000 elementary, middle and high school students from across the U.S. participated in the study. Rasch modeling was used to estimate item and student measures. Option probability curves were used to represent the distribution of correct answers and misconceptions across the range of student achievement levels. Analysis of the shapes of the curves and where they occur as a function of student achievement provides insights on changes in students' thinking as they learn more about a topic. These data can then be used to inform instruction.