Concept Inventories are written to assess undergraduate student understanding of “big picture” ideas in the sciences, mathematics, and engineering. However, the inferences about learning that student responses to Concept Inventories can support have not been systematically studied. In this project, we will examine the relationship between student performance on a rigorously developed Concept Inventory, the Genetics Concept Assessment, relative to performance on newly developed open-response items, and instructor developed end-of-course exams. Our work will help instructors to think about the quality of their local assessment items and the relationship between those items and the Genetics Concept Assessment, and will inform the development and validation of new Concept Inventories across all disciplines. This two-year project sponsored by the W.M. Keck Foundation will launch this 2015-2016 school year as a collaborative project between CADRE, CU Boulder's Department of Molecular, Cellular and Developmental Biology, and the University of Colorado at Denver's School of Education & Human Development.