Students now have readily available and powerful tools to access, manipulate, combine, and visualize data. Acquiring data and visual literacy requires more than knowledge of how to use these tools. Students need to engage with assignments that challenge them to make relatively complex visualizations, interpret them, and explain why these interpretations matter for given problem situations. This paper describes how to structure feedback for these assignments. The few published visualization evaluation rubrics are mainly oriented toward how-to-do-it heuristics. This paper makes a contribution by presenting, defining, and giving examples of the use of an innovative compact rubric for evaluating visualizations (CRVE). This rubric eliminates some of the length and complexity of heuristic evaluation, focusing on interpretation and relevance. In a graduate business intelligence course, students showed definite improvement over the course of the semester in the construction of visualizations, telling a story with headlines, and striving for data exploration. However, higher levels of technical correctness of visualizations did not necessarily correspond to better interpretations. This finding underscores the importance of emphasizing interpretation through a feedback mechanism like the CRVE presented here.