The Additive Factor Model (AFM) is a cognitive diagnostic model that can be used to predict student performance on items in a context that allows for student learning. Within AFM, "skills" have a learning rate, and student acquisition of a skill depends only on the number of opportunities a student has had to exercise that skill and the learning rate of that skill. Here we demonstrate an approach to measure the "teaching value" of individual "items" with respect to one another. The teaching values estimated through this approach may be useful for structuring intelligent tutoring systems and for content improvement. [For the full proceedings, see ED599096.]